INDEX

Explanations

intelligence and potential

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

‖</

2.89

ước

2.83

coprotein

2.64

...</

2.58

 xét

2.55

ObjectMapper

2.54

−</

2.53

">→</

2.51

दीश

2.50

רות

2.47

POSITIVE LOGITS

affin

2.57

меча

2.50

uros

2.37

lectual

2.35

িরা

2.32

紝

2.31

uating

2.30

 stalks

2.26

enial

2.22

 सबै

2.20

Activations Density 0.172%

No Known Activations